Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from
Original research Novel bacterial vaginosis-associated organisms mediate the relationship between vaginal douching and pelvic inflammatory disease Tamala Gondwe ,1 Roberta Ness,2 Patricia A Totten,3 Sabina Astete,3 Gong Tang,4 Melanie A Gold,5 David Martin,6 Catherine L Haggerty 1,7
►► Additional material is ABSTRACT PID, including infertility, ectopic pregnancy, recur- published online only. To view Objectives We sought to determine whether the rent PID and chronic pelvic pain.2 please visit the journal online relationship between a history of vaginal douching Ness et al previously identified an associa- (http://dx. doi.org/ 10.1136/ 3 sextrans-2019- 054191). and pelvic inflammatory disease (PID) is mediated by tion between vaginal douching and PID. Among endometrial infection with one or more novel bacterial women in the PID Evaluation and Clinical Health For numbered affiliations see vaginosis (BV)-associated organisms among Atopobium (PEACH) study, those who reported douching more end of article. vaginae, the BV-associated bacterium 1 (BVAB1), neathia than once per month were 60% more likely to have (Leptotrichia) amnionii and Sneathia sanguinegens. histologically confirmed endometritis or upper Correspondence to 3 Dr Catherine L Haggerty, Methods We first conducted log-binomial regression genital tract gonorrheal or chlamydial infection. Department of Epidemiology, analyses to identify risk factors for endometrial infection Another study in a similar cohort, the Gynecologic University of Pittsburgh in 535 adolescent and adult women with clinically Infections Follow-Through (GIFT) study found that Graduate School of Public suspected PID in the PID Evaluation and Clinical Health douching for symptoms or hygiene at least once per Health, Pittsburgh, Pennsylvania, (PEACH) study. We then examined whether endometrial month was associated with bacterial vaginosis (BV), PA 15260, USA; haggerty@ 4 pitt.edu infection by the BV-associated organisms mediated the diagnosed using Nugent’s criteria. BV is a common association between a history of vaginal douching and condition among women, with prevalence estimates Received 3 July 2019 histologically confirmed PID using inverse probability ranging from 5% to 36%,5 and is characterised by Revised 27 October 2019 weighted marginal structural models. Accepted 7 November 2019 a shift from a vaginal flora with predominant lacto- Published Online First Results Vaginal douching was significantly associated bacilli to one with high concentrations of a diverse 6 December 2019 with endometrial infection with one or more of the collection/array of other aerobic and anaerobic targeted BV-associated organisms (relative risk (RR) bacteria. These findings support the notion that 1.21, 95% CI: 1.08 to 1.35). The total effect estimate douching disrupts the normal vaginal flora, which, suggested that vaginal douching increased the risk of in turn, promotes colonisation with BV-associated endometritis by 24% (RR 1.24, 95% CI: 1.03 to 1.49). bacteria. Women with BV also have an increased risk The controlled direct effect of this association was of acquiring sexually transmitted infections (STIs), http://sti.bmj.com/ attenuated with endometrial infection by one or more particularly Neisseria gonorrhoeae and Chlamydia BV-associated organisms (adjusted RR (aRR) 1.00, trachomatis, which if left untreated can ascend to 95% CI: 0.57 to 1.74) and endometrial infection by all the upper genital tract and cause PID.6 Although four BV-associated organisms (aRR 0.95, 95% CI: 0.53 to PID is often associated with N. gonorrhoeae and 1.70) as intermediate variables. C. trachomatis infection, up to 70% of PID cases Conclusions Endometrial infection with one or more of are of unknown microbial aetiology, and BV has on September 24, 2021 by guest. Protected copyright. the novel BV-associated organisms partially mediated the been directly associated with PID, independent of relationship between vaginal douching and histologically chlamydial and gonococcal infection.7 confirmed endometritis in the EACHP study. Frequent Using traditional bacterial cultivation methods, vaginal douching may confer risk for endometritis the vaginal flora of women with BV is frequently through increasing the risk of endometrial infection defined by overgrowth of Gardnerella vaginalis, by novel-BV -associated organisms. Other potential Ureaplasma species, Mycoplasma hominis, Mobi- pathways should be explored. luncus species, Prevotella species and several other bacteria, including anaerobes.8 The role of recently identified fastidious vaginal microbes in gyneco- Introduction logical and reproductive morbidity has not been © Author(s) (or their Pelvic inflammatory disease (PID), the infection and widely researched. Fredericks et al conducted a employer(s)) 2020. Re-use broad range PCR sequencing study among women permitted under CC BY-NC. No inflammation of a woman’s fallopian tubes (salpin- commercial re-use . See rights gitis) and uterine lining (endometritis), is a frequent aged 18–42 years to identify novel bacteria in and permissions. Published and morbid condition among young women. The vaginal fluid obtained from 27 women with BV and by BMJ. estimated prevalence of PID among sexually experi- 46 women without BV.8 They found a strong associ- To cite: Gondwe T, enced women aged 18–44 years is 4.4%, suggesting ation between BV and Atopobium vaginae, Sneathia Ness R, Totten PA, that approximately 2.5 million reproductive-aged (Leptotrichia) sanguinegens, Sneathia (Leptotrichia) et al. Sex Transm Infect women have ever been diagnosed with PID.1 Major amnionii, and three newly recognised bacteria 2020;96:439–444. reproductive and gynecologic sequelae result from designated Bacterial Vaginosis-associated bacteria
Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191 439 Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from
Figure 1 Illustration of the hypothesised associated pathways in the relationship between a history of vaginal douching twice or more in the past month and endometritis in the PEACH study. PEACH, PID Evaluation and Clinical Health; BV, bacterial vaginosis.
(BVAB): BVAB1, BVAB2 and BVAB3. Furthermore, these bacteria sexual activity in the past 4 weeks, number of lifetime sexual were highly prevalent among women with BV compared with partners, new sexual partner in the past 4 weeks, history of STIs, women without BV. history of BV, history of PID, recent hormonal contraceptive use, We previously demonstrated that vaginal douching is associ- condom use, oral sex, anal sex, age at sexual debut, history of ated with an increased likelihood of cervical and/or endometrial vaginal douching practices, drug and alcohol use, and tobacco infection with S. sanguinegens and S. amnionii in the PEACH use. Vaginal douching at baseline was assessed through questions study, and that endometrial infection by the microorganisms is of douching frequency in the past 4 weeks, type of products used associated with an increased likelihood for endometritis.9 We and reasons for douching. This analysis, consistent with prior suspected that endometrial infection by BV-associated organisms studies, assessed vaginal douching as frequency two times or is an intermediate between vaginal douching and endometritis more in the past month (Yes or No). (figure 1). Our objective was to assess the mediation effect of At baseline and at 30 days after enrolment in the parent PEACH endometrial infection by BV-associated organisms in the rela- study, cervical swabs and endometrial biopsies were obtained for tionship between a history of vaginal douching and histologically histology, chlamydial PCR (Roche Diagnostics) and gonococcal confirmed endometritis using data collected from the PEACH culture. Endometrial biopsies were obtained using a sterile, single cohort. use, disposable endometrial suction curette after first swabbing the exocervix twice with betadine and the endocervix once with Methods betadine. Histological examination was performed on hematox- http://sti.bmj.com/ Study population ylin and eosin-stained and methyl green pyronine stained biopsy We conducted this study using baseline data from the PEACH tissue specimens, read separately by two reference pathologists study. Details of subject recruitment and data collection have who reviewed the slide together to reach consensus on disagree- previously been described.10 In brief, the primary aim of the ment. A modification of the criteria proposed by Kiviat et al was PEACH study was to compare the effectiveness of inpatient used to diagnose endometritis, defined by at least five neutrophils versus outpatient treatment for preventing long-term complica- per ×400 in the endometrial surface epithelium in the absence on September 24, 2021 by guest. Protected copyright. tions of mild-to- moderate PID. In all, 831 women, aged 14–37 of menstrual endometrium and/or at least two plasma cells per years with clinically suspected PID, were recruited from emer- ×120 in the endometrial stroma. For this current substudy, gency departments, OB/GYN clinics, STI clinics and private archived endometrial biopsy and cervical swab specimens were practices in 13 sites throughout the Eastern, Southern and thawed, purified using the MasterPure DNA purification kit for Central regions of the USA. Recruitment for the study took patient specimens (Epicentre) and then tested by species-specific place between March 1996 and February 1999, and the average PCR assays: S. sanguinegens, S. amnionii, A. vaginae and BVAB1, 9 follow-up time was 84 months by the conclusion of the study in as previously described. 2004 (69% participant retention). The University of Pittsburgh Institutional Review Board approved this study. Statistical analysis We restricted analysis to women in the PEACH cohort who Data collection and laboratory analyses had complete endometrial biopsy and PCR assay baseline data Baseline data were collected by trained research staff at each (n=535). Study participants who did not have complete endome- study centre using standardised interviews, examinations and trial biopsy and PCR assay baseline data were more likely to be specimen collection techniques.10 Information was collected African-American (36.6% vs 21.8% Caucasian/Hispanic/Other on demographic characteristics, sexual history and behavioural race, p=0.0001) and report being a current tobacco smoker characteristics. Demographic characteristics evaluated as poten- (36.5% vs 29.6% non-smokers, p=0.0385). In previous analysis tial risk factors for BV-associated bacterial infection included age, of this cohort, endometrial detection of these four BV-associ- race/ethnicity, marital status, employment, health insurance,and ated organisms, A. vaginae, BVAB1, S. amnionii and S. sanguine- level of education. In addition, we assessed reproductive history, gens, were highly correlated with each other, and this microbial
440 Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191 Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from community was predictive of subsequent infertility.9 Therefore, 1.56) and having more than seven alcoholic drinks in the past we similarly combined PCR results for these four BV-associated week (RR 1.19, 95% CI: 1.03 to 1.39) were associated with an bacteria into one variable that indicated endometrial infection increased risk of endometrial infection (table 1). with one or more organism, and a variable that indicated infec- History of douching was also associated with an increased risk tion with all four bacteria. We then conducted univariate anal- of endometrial infection with one or more of the BV-associated yses to explore a range of characteristics and risk factors for organisms (RR 1.21, 95% CI: 1.08 to 1.35) and endometrial endometrial infection with one or more BV-associated organ- infection with all four organisms combined (RR 1.75, 95% CI: isms. Lastly, we assessed the unadjusted risk of endometritis in 1.05 to 2.93) (table 2). In strata of douching frequency, the risk of women infected with one or more BV-associated organisms and endometritis was almost double in those who reported douching women infected with all four organisms, stratified by douching twice or more a month and who had endometrial infection with frequency. one or more or with all of the BV-associated organisms, although Figure 1 depicts the pathways of the proposed mediation. the estimates were not statistically significant (RR 1.93, 95% CI: We used the counterfactual outcomes framework to assess the 0.97 to 3.82). Conversely, in strata of endometrial infection with mediation effect of the BV-associated organisms (M) in the rela- BV-associated organisms, there was a decreased magnitude for tionship between a history of vaginal douching and endometritis risk of endometritis among women with no endometrial infec- using inverse probability weighted marginal structural models tion who reported douching twice or more in a month (RR 0.97, (online supplementary appendix).11 This allowed for estima- 95% CI 0.54 to 1.75) (online supplementary table 1). tion of the controlled direct effect of a history of douching on Report of douching twice or more per month at baseline was endometritis.11 The controlled direct effect is an estimate of the associated with a 24% increased risk of histologically confirmed relative risk of endometritis among women who douched twice endometritis in crude analysis (RR 1.24, 95% CI: 1.03 to 1.49) or more a month if all women were not infected with the BV-as - (table 3). The controlled direct effect of vaginal douching on sociated organisms at baseline (M=0), compared with those endometritis was nullified after removing the intermediate effect who did not, and adjusting for exposure-outcome (CXY) and of endometrial infection with one or more of the bacteria (aRR mediator-outcome confounders (CMY) and accounting for inter- 1.00, 95% CI: 0.57 to 1.74). The findings were similar after action between the exposure (douching) and mediator (BV-as- removing the intermediate effect of endometrial infection with sociated organisms). Confounders were assessed by whether all bacteria (aRR 0.95, 95% CI: 0.53 to 1.70). they changed the estimate by >10% when added to the bivar- In the sensitivity analysis among 273 women who tested nega- iate model predicting the outcome. Based on this method, age tive for N. gonorrhoeae and C. trachomatis infection at baseline, (<25 years vs ≥25 years) and race (African-American vs Cauca- women who reported douching twice or more per month at base- sian/Hispanic/Other) were identified as confounders between line had an approximately 60% increased risk of histologically a history of douching and endometritis, and race was also a confirmed endometritis (RR 1.59, 95% CI: 1.06 to 2.37) (data confounder between endometrial infection with the BV-associ - not shown). This association was nullified after removing the ated organisms and endometritis. intermediate effect of endometrial infection by one or more of The outcome variable was dichotomised into ‘Positive’ or the BV-associated organisms (aRR 0.91, 95% CI: 0.38 to 2.17). ‘Negative’ for histologically confirmed endometritis at baseline. Log-binomial regression models were used to predict the expo- sure (douching) and mediator (BV-associated organisms). The Discussion inverse of the predicted probabilities from the fitted model was In our study of 535 women with clinically suspected PID, we http://sti.bmj.com/ used as weights in subsequent mediation analyses. The inverse found that infection of the endometrium by A. vaginae, BVAB1, probability weighted log-binomial regression models provided and S. amnionii and/or S. sanguinegens is an intermediate in the adjusted relative risks (aRR) and 95% CI. In addition, we relationship between vaginal douching and endometritis. We conducted a sensitivity analysis to assess the controlled direct also identified participant characteristics and risk factors that are effect of a history of vaginal douching on endometritis inde- correlates of infection by A. vaginae, BVAB1, S. amnionii and/ pendent of endometrial infection with the BV-associated organ- or S. sanguinegens. In particular, endometrial infection by one isms in women who tested negative for N. gonorrhoeae and C. or more of the BV-associated organisms is associated with race, on September 24, 2021 by guest. Protected copyright. trachomatis infection at baseline (n=273). All analyses were marital status, level of education, a new sexual partner in the last conducted using SAS V.9.4 (SAS Institute, Cary, NC, USA) and month, history of anal sex and frequent alcohol consumption. To STATA 15. our knowledge, no previous studies have assessed the mediating effects of novel BV-associated organisms on PID. Only a handful of studies have examined the risk factors for Results novel bacteria associated with BV. No previous studies have Of the 535 women included in the analysis, 47.5% had histo- assessed the risk factors specifically for the bacteria designated logically confirmed endometritis at study baseline. Vaginal as BVAB. In an Australian study assessing 342 women with a douching twice or more a month was reported by 100 (18.8%) range of sexual exposures from two cohorts, A. vaginae was of these participants. The majority of participants had endo- detected in sexually naïve women, but was not found to be metrial infection with one or more of these fastidious bacteria significantly associated with high sexual exposure.12 On the (358/535, 66.9%). In univariate analysis, demographic factors other hand, Sneathia (formerly Leptotrichia) spp. was signifi- significantly associated with endometrial infection with one or cantly associated with women who engage in sex with women more BV-associated organisms included African-American race (WSW), and had a strong relationship with increased sexual (RR 1.45, 95% CI: 1.23 to 1.71), being unmarried (RR 1.46, exposure. In contrast to the findings of this study, none of 95% CI: 1.11 to 1.92) and having less than a high school educa- the sexual exposure predictors were statistically significant in tion (RR 1.23, 95% CI: 1.09 to 1.38). In addition, report of univariate and multivariate analyses for endometrial infection having a new sexual partner in the last month (RR 1.18, 95% CI: with one or more BV-associated organism. In addition, the 1.01 to 1.37), history of anal sex (RR 1.26, 95% CI: 1.01 to PEACH study did not assess the prevalence of WSW in this
Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191 441 Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from
Table 1 Characteristics of study participants and association with Table 1 Continued endometrial infection with one or more novel BV-associated organisms Total Positive * Negative in the PEACH study n=535 n=358 n=177 Total Positive * Negative Characteristic N (%) % % RR (95% CI) n=535 n=358 n=177 Age at sexual debut Characteristic N (%) % % RR (95% CI) ≤15 years 277 (52.0) 53.1 49.7 Ref. Demographic >15 years 256 (48.0) 46.9 50.3 0.96 (0.85 to 1.08) Age Risky sexual <25 years 350 (65.4) 65.9 64.4 1.02 (0.90 to 1.16) behaviour** ≥25 years 185 (34.6) 34.1 35.6 Ref. Yes 404 (90.0) 90.6 88.7 1.07 (0.85 to 1.36) Race/ethnicity No 45 (10.0) 9.4 11.3 Ref. African-American 374 (69.9) 77.1 55.4 1.45 (1.23 to 1.71)† Behavioural Caucasian/ 161 (30.1) 22.9 44.6 Ref. Vaginal douche ≥2 Hispanic/Other times in past month Marital status Yes 100 (18.8) 21.9 12.4 1.21 (1.07 to 1.38)† Unmarried 439 (88.2) 91.6 81.3 1.46 (1.11 to 1.92)† No 433 (81.2) 78.1 87.6 Ref. Married 59 (11.8) 8.4 18.7 Ref. Illicit drug use Education Yes 143 (26.9 28.2 24.3 1.07 (0.94 to 1.21)
History of BV n=34; age at sexual debut, n=2; risky sexual behaviour, n=86; vaginal douching, http://sti.bmj.com/ n=2; drug use, n=3; smoking, n=2, alcohol use, n=2; risky social behaviour, n=2. Yes 125 (24.0) 24.4 23.1 1.02 (0.89 to 1.18) *Tested positive for Atopobium vaginae, BVAB1, Sneathia amnionii, or Sneathia No 396 (76.0) 75.6 76.9 Ref. sanguinegens. History of PID †P value<0.05. Yes 163 (30.7) 30.4 31.4 0.98 (0.86 to 1.12) ‡History of N. gonorrhoeae, C. trachomatis, or Trichonomas vaginalis. No 367 (69.3) 69.6 68.6 Ref. §Condoms used 0 to 5 out of 10 sexual encounters. ¶Condoms used 10 out of 10 sexual encounters. Recent hormonal on September 24, 2021 by guest. Protected copyright. **≥2 sexual partners, new sexual partner in the last month, inconsistent condom contraception use use. Yes 98 (21.8) 20.5 24.5 0.92 (0.78 to 1.09) ††Current tobacco smoker, >7 alcohol drinks per week, illicit drug use. No 351 (78.2) 79.5 75.5 Ref. BV, bacterial vaginosis; PEACH, PID Evaluation and Clinical Health; PID, pelvic Rare/occasional inflammatory disease; STI, sexually transmitted infection. condom use§ Yes 318 (70.8) 71.5 69.5 1.03 (0.89 to 1.20) No 131 (29.2) 28.5 30.5 Ref. cohort; therefore, we could not evaluate this as a risk factor in Consistent condom our analysis. use¶ A history of vaginal douching has been previously associated Yes 58 (12.9) 12.4 13.9 0.96 (0.78 to 1.17) with development of PID in the PEACH study,3 but not in the 6 No 391 (87.1) 87.6 86.1 Ref. GIFT study, which was a prospective cohort study. The current History of oral sex study suggests that the observed relationship in the PEACH Yes 121 (24.2) 21.9 28.6 0.88 (0.75 to 1.03) study could in part be mediated by endometrial infection by No 380 (75.8) 78.1 71.4 Ref. novel BV-associated organisms. Vaginal douching is a custom in History of anal sex the USA and around the world that is associated with cultural Yes 18 (3.4) 4.2 1.7 1.26 (1.01 to 1.56)† beliefs about cleanliness commonly inherited by women from their own mothers.13–15 In addition to vaginal douching, it has No 517 (96.6) 95.8 98.3 Ref. been found that women in the USA partake in vaginal inser- Continued tion practices with various other products, such as petroleum
442 Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191 Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from
associated with changes in the microbial composition. Further Table 2 Associations between vaginal douching, endometrial research is needed to assess the effect of douching cessation and infection with BV-associated organisms1 and endometritis in the vaginal bacterial outcomes. PEACH study Strengths of our study include the large study sample size and Histologically confirmed endometritis the objective measure of the mediator and outcome using labo- Douching <2 times / Douching ≥2 times/ ratory techniques, which reduce the chance of misclassification month month error. In addition, we were able to apply counterfactual media- n=433 n=100 tion methods to determine controlled direct effects and adjust RR (95% CI) RR (95% CI) for mediator-outcome confounders.11 Our study is limited by the Endometrial infection with one or cross-sectional design of our analysis, which did not allow us more BV-associated organisms* to prove a causal association. Prospective studies are needed to Yes 1.36 (1.07 to 1.71) 1.622 (0.91 to 2.61) determine the temporal relationship between vaginal douching, No Ref. Ref. BV-associated organisms and PID. Furthermore, as vaginal Endometrial infection with all douching is measured by self-report, it is possible that women four BV-associated organisms* may have under-reported this behaviour, thus resulting in under- Yes 1.53 (1.07 to 2.17) 1.93 (0.97 to 3.82) estimation of the total effect of vaginal douching on endome- No Ref. Ref. tritis. In addition, given that the PEACH study concluded in *BV-associated organisms: Atopobium vaginae, BVAB1, Sneathia amnionii, or 2004, it is possible that the microbial aetiology of PID may Sneathia sanguinegens. have evolved. Lastly, although women in the PEACH study are BV, bacterial vaginosis; PEACH, PID Evaluation and Clinical Health; RR, relative risk. generally representative of all women with PID in terms of risk profile, all PEACH study participants had clinically suspected PID at enrolment, which may limit the generalisability of our jelly, for hygiene and lubrication, and these insertions practices 16 findings. also increase the risk of BV. The rate of vaginal douching in Our study sought to determine whether novel BV-asso- the USA is estimated at 32.2% in women aged 15–44 years, ciated organisms are mediators in the relationship between and is highest among African-American women, followed by vaginal douching and histologically confirmed endometritis in Hispanic women, with Caucasian women reporting the lowest 17 the PEACH study. To our knowledge, this is the first study to rates. A randomised, controlled trial in the southeastern USA examine the potential mediation effect of novel BV-associated showed that behavioural interventions, individualised counsel- microorganisms on vaginal douching and endometritis. This ling sessions, reduced the practice of vaginal douching by almost effect is important to define to better target PID prevention 50% among young women who had no intention to change their 18 efforts. In particular, vaginal douching cessation would have a behaviour at baseline. Similarly, a literature review of research major impact on reduction of PID based on the premise that on vaginal douching found that healthcare providers have the endometrial infection with one or more BV-associated organism most influence in discouraging vaginal douching practices since 19 results from vaginal douching. Additional research should be they can relay the associated health risks. This was evident in conducted to elucidate other demographic and behavioural interviews of two generations of African-American mothers and risk factors for infection with novel BV-associated organisms daughters in another randomised clinical trial which found that in women. Other than vaginal douching and sexual exposure, vaginal douching was less frequent among daughters than among little else is known about the aetiology of these novel organ- mothers, in part due to healthcare provider messages, which http://sti.bmj.com/ isms. Furthermore, optimal treatment options should be investi- suggests that with increased education and awareness, the trend gated in light of drug resistance to commonly used therapies. For of vaginal douching acceptance is changing and can potentially be 17 instance, the A. vaginae strains tested to date have been resistant eliminated. Our findings suggest that behaviour modification to metronidazole, the current first line of treatment for uncom- interventions are warranted to discourage the uptake of vaginal plicated BV infection.20 douching to reduce the risk of acquiring endometrial infection Recurrent BV infection has also been attributed to the ineffectiveness of metronidazole, confirming by BV-associated organisms, and subsequently developing PID. 20 on September 24, 2021 by guest. Protected copyright. In the present study, it is not clear whether practising vaginal the urgency for new and more effective therapies. In conclu- douching leads to a change in the vaginal microbial composi- sion, whether endometrial infection by novel BV-associated tion, or whether women are douching to alleviate symptoms organisms is on the causal pathway between vaginal douching
Table 3 Crude and controlled-direct associations between vaginal Key messages douching and endometritis in the PEACH study Controlled direct Controlled direct ►► While vaginal douching has been found to be associated with effect accounting for effect accounting for pelvic inflammatory disease, it is unknown whether infection endometrial infection Endometrial infection with novel bacterial vaginosis-associated organisms is on the with one or more BV- with all four BV- Total effect associated organisms* associated organisms* causal pathway. RR aRR† aRR† ►► Using inverse probability weighted marginal structural (95% CI) (95% CI) (95% CI) models allowed for determination of the controlled direct Vaginal douche effect of vaginal douching on endometritis after adjusting for ≥2 times in past month confounders. Yes 1.24 (1.03 to 1.49) 1.00 (0.57 to 1.74) 0.95 (0.53 to 1.70) ►► Endometrial infection with either Atopobium vaginae, No Ref. Ref. Ref. BVAB1, Sneathia amnionii and/or Sneathia sanguinegens was
*BV-associated organisms: Atopobium vaginae, BVAB1, Sneathia amnionii, or Sneathia sanguinegens. found to partially mediate the association between vaginal †Adjusted for race and age. douching and endometritis. BV, bacterial vaginosis; PEACH, PID Evaluation and Clinical Health.
Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191 443 Epidemiology Sex Transm Infect: first published as 10.1136/sextrans-2019-054191 on 6 December 2019. Downloaded from and endometritis warrants further research to provide impetus Catherine L Haggerty http://orcid. org/0000- 0002-9849- 7865 on the part of healthcare providers to (1) strongly discourage the initiation of vaginal douching, particularly among women References of reproductive age and especially adolescents and (2) optimise 1 Kreisel K, Torrone E, Bernstein K, et al. Prevalence of Pelvic Inflammatory Disease screening for and treatment of novel BV-associated organisms to in Sexually Experienced Women of Reproductive Age - United States, 2013-2014. prevent the adverse sequelae of PID. MMWR Morb Mortal Wkly Rep 2017;66:80–3. 2 Weström L, Joesoef R, Reynolds G, et al. Pelvic inflammatory disease and fertility.A cohort study of 1,844 women with laparoscopically verified disease and 657 control Author affiliations women with normal laparoscopic results. Sex Transm Dis 1992;19:185–92. 1 Department of Epidemiology, University of Pittsburgh Graduate School of Public 3 ness RB, Soper DE, Holley RL, et al. Douching and endometritis: results from the PID Health, Pittsburgh, Pennsylvania, USA evaluation and clinical health (peach) study. Sex Transm Dis 2001;28:240–5. 2 Department of Epidemiology, Human Genetics & Environmental Sciences, The 4 ness RB, Hillier SL, Richter HE, et al. Douching in relation to bacterial vaginosis, University of Texas School of Public Health, Houston, Texas, USA lactobacilli, and facultative bacteria in the vagina. Obstet Gynecol 2002;100:765. 3 Department of Medicine, Division of Infectious Diseases, University of Washington, 5 Morris M, Nicoll A, Simms I, et al. Bacterial vaginosis: a public health review. BJOG Seattle, Washington, USA 2001;108:439–50. 4 Department of Biostatistics, University of Pittsburgh Graduate School of Public 6 ness RB, Hillier SL, Kip KE, et al. Douching, pelvic inflammatory disease, and incident Health, Pittsburgh, Pennsylvania, USA gonococcal and chlamydial genital infection in a cohort of high-risk women. Am J 5 Division of Child and Adolescent Health, Department of Pediatrics, Columbia Epidemiol 2005;161:186–95. University Irving Medical Center and Department of Population and Family Health, 7 haggerty CL, Hillier SL, Bass DC, et al. Bacterial vaginosis and anaerobic bacteria are Mailman School of Public Health, Columbia University, New York, New York, USA associated with endometritis. Clin Infect Dis 2004;39:990–5. 6 Department of Medicine, Louisiana State University, New Orleans, Louisiana, USA 8 Fredricks DN, Fiedler TL, Marrazzo JM. Molecular identification of bacteria associated 7 Magee- Womens Research Institute, Pittsburgh, Pennsylvania, USA with bacterial vaginosis. N Engl J Med 2005;353:1899–911. 9 haggerty CL, Totten PA, Tang G, et al. Identification of novel microbes associated with Handling editor Anna Maria Geretti pelvic inflammatory disease and infertility. Sex Transm Infect 2016;92:441–6. 10 ness RB, Soper DE, Peipert J, et al. Design of the PID evaluation and clinical health Acknowledgements The authors acknowledge Dr Michael Ferris and Johana (peach) study. Control Clin Trials 1998;19:499–514. Norori for their contribution to the PEACH study. 11 VanderWeele TJ. Marginal structural models for the estimation of direct and indirect Contributors All authors have seen and approved the manuscript, have effects. Epidemiology 2009;20:18–26. contributed significantly to the work, and do not have conflicts of interest to report. 12 Fethers K, Twin J, Fairley CK, et al. Bacterial vaginosis (bv) candidate bacteria: Funding This work was supported by the National Institute of Allergy and associations with bv and behavioural practices in sexually-experienced and Infectious Diseases at the National Institutes of Health [R01AI073940 to C.L.H.] inexperienced women. PLoS One 2012;7:e30633. 13 rupp R, Short MB, Head-Carroll Y, et al. Intergenerational transfer of douching Competing interests None declared. information. J Pediatr Adolesc Gynecol 2006;19:69–73. Patient consent for publication Not required. 14 ness RB, Hillier SL, Richter HE, et al. Why women douche and why they may or may not stop. Sex Transm Dis 2003;30:71–4. Ethics approval University of Pittsburgh Institutional Review Board (IRB number: 15 rajamanoharan S, Low N, Jones SB, et al. Bacterial vaginosis, ethnicity, and the use of 0608052). genital cleaning agents: a case control study. Sex Transm Dis 1999;26:404–9. Provenance and peer review Not commissioned; internally peer reviewed. 16 Brown JM, Hess KL, Brown S, et al. Intravaginal practices and risk of bacterial vaginosis and candidiasis infection among a cohort of women in the United States. Data availability statement Data are available upon reasonable request with Obstet Gynecol 2013;121:773–80. approval from the corresponding author at the contact provided. 17 Mark H, Sherman SG, Nanda J, et al. What has changed about vaginal Open access This is an open access article distributed in accordance with the douching among African American mothers and daughters? Public Health Nurs Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which 2010;27:418–24. permits others to distribute, remix, adapt, build upon this work non-commercially, 18 grimley DM, Oh MK, Desmond RA, et al. An intervention to reduce vaginal douching and license their derivative works on different terms, provided the original work is among adolescent and young adult women: a randomized, controlled trial. Sex Transm properly cited, appropriate credit is given, any changes made indicated, and the use Dis 2005;32:752–8. http://sti.bmj.com/ is non-commercial. See: http://creativecommons.org/ licenses/by- nc/4. 0/. 19 cottrell BH. An updated review of of evidence to discourage douching. MCN Am J Matern Child Nurs 2010;35:102–7. ORCID iDs 20 sobel R, Sobel JD. Metronidazole for the treatment of vaginal infections. Expert Opin Tamala Gondwe http://orcid. org/0000- 0002-9288- 2113 Pharmacother 2015;16:1109–15. on September 24, 2021 by guest. Protected copyright.
444 Gondwe T, et al. Sex Transm Infect 2020;96:439–444. doi:10.1136/sextrans-2019-054191